dacorvo HF Staff commited on
Commit
1095503
·
verified ·
1 Parent(s): e87bae8

Synchronizing local compiler cache.

Browse files
.gitattributes CHANGED
@@ -11569,3 +11569,6 @@ neuronxcc-2.21.18209.0+043b1bf7/MODULE_6a4730946575bae38107+a9d440f5/wrapped_nef
11569
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_154c082decaad50606bc+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11570
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_154c082decaad50606bc+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11571
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_88724518ccebd3fa6470+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
 
 
 
 
11569
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_154c082decaad50606bc+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11570
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_154c082decaad50606bc+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
11571
  neuronxcc-2.21.18209.0+043b1bf7/MODULE_88724518ccebd3fa6470+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11572
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_c243c517d60e8dd22317+ed72d204/model.neff filter=lfs diff=lfs merge=lfs -text
11573
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/model.neff filter=lfs diff=lfs merge=lfs -text
11574
+ neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/wrapped_neff.hlo filter=lfs diff=lfs merge=lfs -text
neuronxcc-2.21.18209.0+043b1bf7/0_REGISTRY/0.4.0.dev0/phi3/microsoft/phi-4/f68b397ab9d0ed2bd338.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_entry_class": "SingleModelCacheEntry",
3
+ "_model_id": "microsoft/phi-4",
4
+ "_task": "text-generation",
5
+ "architectures": [
6
+ "Phi3ForCausalLM"
7
+ ],
8
+ "attention_bias": false,
9
+ "attention_dropout": 0.0,
10
+ "embd_pdrop": 0.0,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 5120,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 17920,
15
+ "max_position_embeddings": 16384,
16
+ "model_type": "phi3",
17
+ "neuron": {
18
+ "_serialized_key": "NxDNeuronConfig",
19
+ "batch_size": 1,
20
+ "capacity_factor": null,
21
+ "checkpoint_id": "microsoft/phi-4",
22
+ "checkpoint_revision": "187ef0342fff0eb3333be9f00389385e95ef0b61",
23
+ "continuous_batching": false,
24
+ "enable_bucketing": false,
25
+ "ep_degree": 1,
26
+ "fused_qkv": true,
27
+ "glu_mlp": true,
28
+ "local_ranks_size": 10,
29
+ "max_batch_size": 1,
30
+ "max_context_length": 4096,
31
+ "max_topk": 256,
32
+ "n_active_tokens": 4096,
33
+ "neuronxcc_version": "2.21.18209.0+043b1bf7",
34
+ "on_device_sampling": true,
35
+ "optimum_neuron_version": "0.4.0.dev0",
36
+ "output_logits": false,
37
+ "pp_degree": 1,
38
+ "sequence_length": 4096,
39
+ "speculation_length": 0,
40
+ "start_rank_id": 0,
41
+ "target": "trn1",
42
+ "torch_dtype": "bfloat16",
43
+ "tp_degree": 10
44
+ },
45
+ "num_attention_heads": 40,
46
+ "num_hidden_layers": 40,
47
+ "num_key_value_heads": 10,
48
+ "original_max_position_embeddings": 16384,
49
+ "partial_rotary_factor": 1.0,
50
+ "resid_pdrop": 0.0,
51
+ "rms_norm_eps": 1e-05,
52
+ "rope_scaling": null,
53
+ "rope_theta": 250000,
54
+ "sliding_window": null,
55
+ "tie_word_embeddings": false,
56
+ "use_cache": true,
57
+ "vocab_size": 100352
58
+ }
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c243c517d60e8dd22317+ed72d204/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/context_encoding_model/_tp0_bk0/log-neuron-cc.txt"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c243c517d60e8dd22317+ed72d204/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c243c517d60e8dd22317+ed72d204/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:268ec4b3b176121a5ae8b25bcdc88611c35c238652394bd18c1e322310a9ea16
3
+ size 1141233
neuronxcc-2.21.18209.0+043b1bf7/MODULE_c243c517d60e8dd22317+ed72d204/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:441bd2467d1eadead83cfa2ed49e612ad954bbb6be5b2b50a26c9ce9c786dc31
3
+ size 2632704
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/compile_flags.json ADDED
@@ -0,0 +1 @@
 
 
1
+ ["--target=trn1", "--auto-cast=none", "--model-type=transformer", "--tensorizer-options=--enable-ccop-compute-overlap --cc-pipeline-tiling-factor=2 --vectorize-strided-dma ", "-O2", "--lnc=1", "--logfile=/tmp/nxd_model/token_generation_model/_tp0_bk0/log-neuron-cc.txt", "--enable-internal-neff-wrapper"]
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/model.done ADDED
File without changes
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/model.hlo_module.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:997fce42d5fb854e9a2272d02f75583315792f09ae6cc56a044ddaebffd39bb6
3
+ size 1002831
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/model.neff ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce47a31bd56247cd790d6658727a1518f5c3b0ddcd2aae37d677bae36020144f
3
+ size 3810304
neuronxcc-2.21.18209.0+043b1bf7/MODULE_ea0b08427c20ef00b9e3+a9d440f5/wrapped_neff.hlo ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a7967d6022e0ddd522b9fa5066b3d53796779a133d440ca96422f542bdeee53
3
+ size 3995906